---
license: apache-2.0
base_model: microsoft/swinv2-tiny-patch4-window8-256
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: swinv2-tiny-patch4-window8-256-OT
  results:
  - task:
      name: Image Classification
      type: image-classification
    dataset:
      name: imagefolder
      type: imagefolder
      config: default
      split: validation
      args: default
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.8225806451612904
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# swinv2-tiny-patch4-window8-256-OT

This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on the imagefolder dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6192
- Accuracy: 0.8226

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.00015
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 40

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| No log        | 0.91  | 5    | 8.8439          | 0.0806   |
| 8.7922        | 2.0   | 11   | 8.0016          | 0.0806   |
| 8.7922        | 2.91  | 16   | 6.0009          | 0.0806   |
| 6.5264        | 4.0   | 22   | 2.7431          | 0.0806   |
| 6.5264        | 4.91  | 27   | 1.3018          | 0.4516   |
| 2.16          | 6.0   | 33   | 1.2696          | 0.4516   |
| 2.16          | 6.91  | 38   | 1.2057          | 0.4516   |
| 1.2876        | 8.0   | 44   | 1.2157          | 0.4516   |
| 1.2876        | 8.91  | 49   | 1.2459          | 0.4516   |
| 1.2456        | 10.0  | 55   | 1.2110          | 0.4516   |
| 1.1901        | 10.91 | 60   | 1.1861          | 0.4516   |
| 1.1901        | 12.0  | 66   | 1.0847          | 0.4677   |
| 1.0665        | 12.91 | 71   | 1.0944          | 0.4677   |
| 1.0665        | 14.0  | 77   | 1.1854          | 0.4677   |
| 1.033         | 14.91 | 82   | 1.0252          | 0.5      |
| 1.033         | 16.0  | 88   | 1.2164          | 0.5161   |
| 1.0323        | 16.91 | 93   | 1.0643          | 0.5      |
| 1.0323        | 18.0  | 99   | 0.9802          | 0.6613   |
| 0.9329        | 18.91 | 104  | 0.9475          | 0.5968   |
| 0.8619        | 20.0  | 110  | 0.9115          | 0.6452   |
| 0.8619        | 20.91 | 115  | 0.8894          | 0.6452   |
| 0.8019        | 22.0  | 121  | 0.8276          | 0.6935   |
| 0.8019        | 22.91 | 126  | 0.8156          | 0.6774   |
| 0.7675        | 24.0  | 132  | 0.7928          | 0.6290   |
| 0.7675        | 24.91 | 137  | 0.7163          | 0.7419   |
| 0.6762        | 26.0  | 143  | 0.7388          | 0.6774   |
| 0.6762        | 26.91 | 148  | 0.6519          | 0.7581   |
| 0.6771        | 28.0  | 154  | 0.6710          | 0.7419   |
| 0.6771        | 28.91 | 159  | 0.6074          | 0.7581   |
| 0.6424        | 30.0  | 165  | 0.6729          | 0.7258   |
| 0.6139        | 30.91 | 170  | 0.5744          | 0.7903   |
| 0.6139        | 32.0  | 176  | 0.6192          | 0.8226   |
| 0.5713        | 32.91 | 181  | 0.6453          | 0.7903   |
| 0.5713        | 34.0  | 187  | 0.6392          | 0.7903   |
| 0.5462        | 34.91 | 192  | 0.5956          | 0.8226   |
| 0.5462        | 36.0  | 198  | 0.5893          | 0.8226   |
| 0.5393        | 36.36 | 200  | 0.5898          | 0.8226   |


### Framework versions

- Transformers 4.36.2
- Pytorch 2.1.2+cu118
- Datasets 2.16.1
- Tokenizers 0.15.0